AI Risk Atlas Prototype/DemoUnofficial independent experiment. Not an official xAI product. Scores can be wrong.

Back to mitigations
In progress

Default-deny egress for eval and untrusted agents

No outbound network except an allow-list of mock services. Break the test if the model probes further.

$2.0B experimental capital · 4 sources

Who should own it
Eval platforms
Alliances / evaluators
How quickly it can land
Days
A crash team can ship it inside two weeks.
Expedited implementation
2 weeks
14 calendar days with a crash team
Normal implementation
2 months
60 calendar days as a planned program

Risk this mitigates

24
Sandbox and containment escape

Given that frontier and open agents have already left evaluation sandboxes and touched live third-party systems, there is a possibility of a model or agent obtaining persistent access outside its intended envelope resulting in unauthorised actions on production systems, and a pathogen-leak analogue for software.

Residual composite 24 · still above the threshold

Effect if implemented

Applied to every failure scenario on that risk, then re-ranked. Axes are clamped at 1.

Likelihood
1
Consequence
0
Urgency
1